This action might not be possible to undo. Are you sure you want to continue?
Contents lists available at ScienceDirect
Decision Support Systems
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / d s s
The complex problem of monetizing virtual electronic social networks☆
Eric K. Clemons
Operations and Information Management, The Wharton School of Business, University of Pennsylvania, 572 Jon M. Huntsman Hall, 3730 Walnut Street, Philadelphia, PA 19104-6340, USA
a r t i c l e
i n f o
a b s t r a c t
As traditional advertising is losing its impact, both advertisers and the media owners who are dependent upon them are desperately seeking alternative ways to reach consumers and alternative ways to earn revenues by doing so. Although there are many ways to earn money from social network trafﬁc, attempting to do so by treating social networks as just another entertainment medium is doomed to failure. Traditional entertainment and advertising media were unidirectional message-sending systems, where paid messages were pushed at viewers as part of the price they paid for being entertained. In contrast, online social networks are better viewed as milieux, or as meeting places where people congregate to exchange information, observe and emulate each other, assess status, and above all entertain themselves. People are not captives, and when they are not interested, are not entertained, or feel that they are being manipulated, they simply leave. The guidelines for monetizing milieux and meeting places are still being discovered. © 2009 Elsevier B.V. All rights reserved.
Available online 15 May 2009 Keywords: Advertising Monetizing online social networks Monetizing the internet Online revenue without advertising Social networks
1. Introduction This paper addresses the recent rush towards using social networks to replace traditional media in the distribution of traditional advertising, and attempts to demonstrate that most recent investments in social websites for this purpose are misguided and overpriced. We do not suggest that social networking websites are without value, merely that it will be extremely difﬁcult to monetize online communities through advertising as if they were no more than alternative media, the natural next step in the evolution from wandering bards, to books and newspapers, and then to radio and television. This paper is written from the perspective of a strategist examining resources and their deployment, potential sources of economic rents, and the defensibility of any potential super-normal rents, and has beneﬁted from the perspective of anthropologist research collaborators and their ethnographic observations. It does not focus on technology or on current usage patterns. Section 2 presents a very short review of networks and social networks. Section 3 presents a review of changes in advertising, Section 4 reviews problems currently faced by advertisers, and Section 5 reviews common
☆ I gratefully acknowledge the enormous contributions of anthropologists Steve Barnett and Arjun Appadurai, who were my coauthors on earlier papers on online social networks. I also wish to thank Professor Frederick Brunel of Boston University, who suggested the metaphor of online social networks as an alternative space or milieu rather than an alternative telecommunications medium, and Professor Cathy Curley, also of Boston University and the Advisory Board of Communispace, for her contributions to my understanding of the role of online communities. Finally, thanks to the anonymous reviewers and editors, who helped me, to clarify the ideas and their presentations. E-mail address: firstname.lastname@example.org. 0167-9236/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.dss.2009.05.003
mistakes and inappropriate responses seen from advertisers facing these problems. It suggests that acquisition of online social networking sites aimed solely at replacing traditional media advertising with social networking advertising will have mixed results, and indeed early evidence is mixed at best. Section 6 reviews the numerous ways in which individuals use social networking websites, while Section 7 examines how these uses may lead to revenue and to economic value for their operators. Section 8 examines the conditions needed for websites to capture trafﬁc, which we have called the new 4 Ps of social networking. Section 9 reviews the issues that appear to create the greatest uncertainty, and conclusions and early predictions are shown in Section 10. 2. Social networks 2.1. Traditional social networks Social networks are familiar to all who study primates, from baboon troops and gorilla and chimpanzee groups to human societies in all levels of cultural development [17,19,20,40]. A social network can be a group of friends living within a city, or a group of college classmates who remain in frequent contact socially. It can also be a group formed speciﬁcally to accomplish a poorly structured set of tasks over time, which then disbands after the tasks are completed, or a support group like a group of pregnant working mothers or cancer sufferers, where older members constantly drop out and are replaced by new entrants. Humans in all cultures at all times form complex social networks; the term social network here means ongoing relations among people that matter to those engaged in the group, either for speciﬁc reasons (like fantasy football, cancer support groups, task forces at work) or for more
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
general expressions of mutual solidarity (like families, clans, friends, social clubs). Social networks are validated by shared perceptions of worth, typically expressed as cultural symbols. Likewise, social networks among individuals who may not be related can be validated and maintained by agreement on objectives, social values, or even by choice of entertainment. Membership in these networks can be relatively permanent (extended families, which endure for lifetimes) or ﬂexible (pregnancy support groups, in which members rotate out after a few months). They involve reciprocal responsibilities and roles that may be altruistic or self-interest based (or a combination or both). Cultural anthropologists have written extensively about these networks, from studies of caste and marriage systems to informal links in the workplace or in the neighborhood. Social networks are trusted because of shared experiences and the perception of shared values or shared needs. For example, new parents will get recommendations from their neighbors for a pediatrician and, over time, for a baby sitter. Friends tell friends about restaurants and movies. It is not yet clear to what extent virtual social networks are trusted or seen as equivalent to traditional networks. Data are conﬂicting, where users in the age groups between 22 and 30 seem to make more extensive use of online social networks, but may also be more resistant to ads that are inserted into the content of the pages that they are viewing.1,2 2.2. Online social networks The newest attempts at the development of social networks have tried to create online social networks without relying upon an initial face-to-face encounter to create the network. Initially, Facebook relied more heavily on face-to-face contact, but increasingly friends are second-order contacts (friends of friends) or higher, or have not yet met (e.g., political action groups, or groups organized to assist in the orientation of freshman before they arrive on campus). In contrast, participants in Second Life, World of WarCraft, Flickr, and YouTube may never have met and may not intend to meet. In this work we distinguish between online support of traditional social networks, such as the use of email to replace telephone calls to schedule a meeting of a monthly book club, and the development of entirely new online communities, as seen in virtual reality websites, gaming websites, and other interactions among individuals who may never actually meet. These entirely new online networks we term virtual electronic social networks, (or ESNs); these networks may have intense interaction, and may achieve signiﬁcant salience in the lives of their participants, but there may be no interaction among members other than within the ESNs. Ever since the internet began to develop online “communities,” there has been active debate about whether online communities actually replicate in crucial ways face-to-face networks. Howard Rheingold in a number of articles and books has been the strongest advocate of the depth and power of online communities as “actual” social networks that correspond in important ways to older forms of networking . Others, like John Perry Barlow, have suggested the opposite — that online communities are at best truncated networks, lacking in depth and intensity. In Barlow's view, participants in online social networks do not have to interact in multiple and complex ways but rather can assume special identities only for online interactions and can also end these interactions more abruptly than in “real life.” For the most part, we agree with Barlow, with the caveat that as internet possibilities expand, the potential for social networks may grow beyond what we see at present. The virtual reality shared world,
Second Life, enables participants to enter a complex world, to choose their avatars (digital manifestation), to build homes and businesses, and to a great extent shape the evolution of that world. Online businesses in Second Life sell prime “real estate,” virtual homes in preferred locations, or sell real music that can be enjoyed ofﬂine. Politicians hold press conferences and meetings. The online interactions in Second Life have morphed into actual businesses, with monetary transactions that can be converted into (real world) US dollars, so that actual money is now a part of a virtual environment.3 Our own experience is that the behavior of individuals in online networks can be very different from the behavior of the same individuals interacting in a more traditional social network. Since its inception over 200 years ago the motto of the London Stock Exchange has been “dictum meum pactum,” (translation: “My word is my bond”) and members observed the highest ethical standards in their face-toface interactions. The members knew whom they were interacting with even after the move to online trading, and yet their behavior towards each other was dramatically different when they abandoned the ﬂoor and its face-to-face interaction . Our focus in the rest of this paper will be on internet social networks and virtual communities, in multiple categories that are now popular with individual users. Examples include Facebook and MySpace, Linkedin, and YouTube. It also includes massively multi-user domains such as the Second Life virtual reality and gamer sites such as World of WarCraft. These have experienced tremendous commercial interest, including a number of high proﬁle and high visibility acquisitions. They are often seen as the next generation of marketing, which we will address below. 2.3. Monetizing electronic social networks A basic premise of this paper is that social networks, whether characterized by physical meetings or online interactions, have their own rules of behavior, which discourage overt attempts at selling. This is not to suggest that no selling is possible in a social network; ads ﬂank the periphery of Flickr's site for sharing photographs (ﬂickr. com), YouTube's site for sharing videos (YouTube.com), and of course TripAdvisor's site for sharing reviews of travel experiences (tripadvisor.com). Mostly, the content on these sites does not attempt to sell anything directly, from one user to another, or even indirectly, with one user recommending products or services to another. Social interactions are not designed to be hijacked and then used for commercial purposes. Readers who doubt this are cordially invited to conduct a simple experiment: (1) Take a social setting, like a class you are teaching, or a dinner party for a friend that you are hosting, or your daughter's graduation party, and (2) select a product or service in which you have a signiﬁcant economic interest, ideally as an investor, or one for which you are receiving a promotional fee, and then (3) try to sell the product or service to the participants at your social gathering. Our experience suggests that this will simply be too embarrassing, both for you and the other participants, for you to attempt this a second time. Based on this premise, we do not consider the large literature on the effectiveness of different approaches to targeting advertising, or the literature on analyzing click-stream data, or the emerging literature on pricing keywords for sponsored search, and the relative merits of pay per impression, pay per click, or pay per sale pricing in sponsored search. The premise of this paper suggests that current forms of advertising will not transfer over to the use of electronic social networks for commercial purposes. If this premise is true, then the large and interesting literature on online advertising has little to
1 This claim is based on the author's ongoing research on social networks and corporate behavior, or social networks and the impact of mobile advertising [10,11,38]. 2 The reader interested in more detail on recent work in social network theory and the systems approach to social networks is referred to [1,5,7,28,35].
3 To see the size of this new economy, its population, and its recent trends and history, visit their information site (secondlife.com/whatis/economy_stats.php).
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
tell us about the commercial use of ESNs. On the other hand, if this premise is false, the paper still offers some additional insights into other, non-advertising sources of revenue. This is not to suggest that the literature on internet marketing, internet advertising, and personalization [2,29,41], or viral marketing and inﬂuentials [25,26,39], or social network marketing [15,31,32] are not helpful in understanding emerging trends in advertising and marketing. Rather, we suggest that they are orthogonal to understanding the role of online social networks, which we argue are difﬁcult to harness or to convert to commercial objectives. 3. Marketing has changed and traditional push-based advertising is dying This section suggests that traditional advertising is losing its credibility and its impact. Simultaneously, consumers are altering their behavior in ways that suggest they are better informed, not more poorly informed. The obvious inference is that consumers are gaining their information from alternative sources, and preliminary evidence suggests that they are getting an increasing amount of the information used to evaluate product and service offerings by consulting other consumers, principally through online reviews and electronic social networking websites. Alternatively, one can argue that at least some of the problems are more structural, dealing with consumers' changing attitudes towards paid commercial messages; if so, this would limit some of the opportunities for online advertising as well . 3.1. Loss of trust in advertising The public distrusts advertising messages and advertising is losing much of its impact. Recent experiments by Ariely conﬁrm the extent to which consumers have come to distrust paid advertising and come to distrust information that is pushed at them from a source that is seen as having an interest in inﬂuencing outcomes.4 This is true whether the provided information comes from a trusted company or a political party. Ariely describes two experiments, both of which are aimed at assessing the extent to which consumers alter their perception of the reliability of the statement based solely on its source. In one of Ariely's experiments subjects were asked to assess the truth value of a large collection of statements; half were indeed true, but consistent with a slight but well known bias towards trust subjects accepted about 55% of the statements. The willingness to accept the statements dropped signiﬁcantly, and the fraction accepted was closer to one third, when the statements were attributed to the Democratic Party, the Republican Party, or to Procter & Gamble. Interestingly, the willingness to accept statements from either party was not signiﬁcantly better among members of that party than among members of the opposition party. Equally interestingly, this lack of trust was evidenced in the rejection of statements that had nothing to do with candidates, with the policy of either party, or with products from P&G or its competitors. In a second experiment, subjects read a review of a stereo system and then listened to the stereo and were asked to rate it. In one case subjects were told that the review was from a highly regarded stereo store close to MIT and in another case they were told that the review was from Consumer Reports. The wording of the review and the listening conditions for the stereo were identical in both cases. Subjects were asked to evaluate and rate the stereo after listening to it. Those who believed the favorable review was from Consumer Reports actually gave the stereo signiﬁcantly higher assessments after listening. Ariely describes the collapse of credibility of paid messages as an example of the “Tragedy of the Commons” . In the traditional tragedy of the commons, all households start by grazing a single cow,
4 This section is based on personal communication and recent and unpublished work by Professor Ariely while he was at MIT's Media Lab. For a more detailed treatment of his work, see .
the commons supports all cows, and every household has enough milk. A single resident decides to graze two cows; the decrease in the welfare of each cow is small, no household suffers signiﬁcantly, and the individual with two cows is almost twice as well off as he was before his addition of the second cow. One by one, each of his neighbors adds a second cow as well, until ultimately the common land is over-grazed, each cow is near starvation and gives very little milk, and everyone is worse off than before. Ariely argues that as long as any single advertiser is tempted to over-claim, over-grazing in the commons of consumers' trust, consumers' trust will be weakened, additional advertisers will be tempted to make exaggerated claims in order to have an impact on consumers' perceptions, and ultimately each advertiser will be forced to over-promise and to over-claim. This will effectively over-graze the commons of consumer trust. 3.2. Organic informedness in place of traditional advertising Consumers increasingly make their ﬁne-tuned purchasing decisions based on their strong individual preferences, and consumers are now far better informed about their options than they were previously. Consumers know what is available, they know what it costs, and they know what they want [12,13]. Our own data and analyses as well as those of other researchers (e.g., [15,31]) support the fact that the increase in consumer informedness is real, that it signiﬁcantly alters consumer behavior, and that its effects are complex and vary by category and vary with the nature of the consumer's objectives. • In categories where consumers seek to avoid bad experiences, it is the absence of negative ratings (rather than the presence of positives) that correlate with sales . • When consumers seek delight and are willing to pay premium prices, it is the strength of the best reviews that predict sales rather than the expected strength of the average ratings . Both effects suggest an increasing role for true awareness of the full range of choices, including prices and product attributes. The available information indicates that informedness alters customers' shopping behavior even for purchases that are made in traditional physical venues. When shopping for cars, the vast majority of consumers perform some type of online comparison shopping before visiting a dealership; this is true even though car purchases are rarely made fully online today. Likewise, our previous studies indicate that informedness alters beer purchases, although for reasons ranging from shipping costs to state regulations very little beer is actually bought or sold online. The correlation between user behavior and the content of speciﬁc online social content reviewing sites suggests that online content contributes to organic informedness, the state of consumers just knowing, without commercial advertising, and knowing in far more detail than is normally available through commercial advertising . This suggests that to a great extent organic informedness is not just augmenting but frequently is replacing paid corporate promotions. 4. The decline of advertising reach, along with decline in credibility There are now more media choices for the viewer. First there were three national network television channels, next came small independent stations. Then cable TV provided still more choice. Now cable subscribers can have access to well over 100 channels; the addition of on demand capability adds hundreds more choices. Rarely does a signiﬁcant portion of a potential market for a product view the same shows, tremendously diluting the impact of media for the delivery of advertising messages. It is widely observed that the TV audience is increasingly fragmented into large numbers of smaller special interest audiences.
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
There are new alternatives to media broadcasts, such as iPods, DVDs, and video games, reducing the amount of time spent watching traditional television broadcasts. • Network TV viewers are down 15% over the last 2 years and this number grows to 22% for younger (14–22) people. Cable TV is more tightly targeted than network TV and ads appeal to smaller numbers of viewers, but viewers who are supposedly more interested. And TV is declining in favor of various podcasts, games, and other online sources of entertainment directly available on PDAs and computers; perhaps this is why the decline is greatest among young males.5 • Some formats of broadcast television have suffered more than others. Broadcast national and local news, both networks and cable, have seen their viewership collapse; after the evening news destroyed the afternoon metropolitan newspapers throughout the United States, they have since largely been replaced by a shift to ondemand internet news. The State of the News Media 2006 (An Annual Report on American Journalism) notes that although morning news also suffered, evening news viewership over all networks declined from approximately 52 million viewers in 1980 to approximately 27 million today.6 The Washington Post reports a similar loss of television viewers to the internet.7 The last two years have been especially brutal, with networks news down from 38% to 30% and local news viewership down from 64% to 56% in just the past two years. • Newspapers continue to lose readers to websites as well.8 • And for some age groups the loss has been even more dramatic. “… for the core video gamer demographic, 18–34 males, TV consumption has declined by roughly 12% while spending 20% more time playing games. And when people do watch TV, they TiVo through the ads, talk on the phone and/or surf the Internet.9” 5. Misinterpreting the data, misplacing the bets It is easy and comforting for advertisers and their agencies to believe that the failure of advertising is caused by technology (e.g., TiVo) or by media fragmentation rather than by consumers' loss of interest in current media choices. Likewise, it is easy and comforting to believe that the failure of advertising is caused by technology, media, and other alternatives rather than a growing distrust of all paid pushbased communications. These mistakes result in simplistic search for alternative media as the easiest response to the belief that it is current media that are failing them, not consumers' unwillingness to accept their message. Media giants, advertisers, and their agencies all do want to believe that the problem is that they are pitching their message via the wrong channels. The belief that the problem is with the media and not with the message is driving advertisers and their agencies into a frenzied search for alternative media. This search for alternative means to deliver the same message does much to explain the very high and potentially risky valuations on Facebook, MySpace, and YouTube: • YouTube is still exploring business models that enable it to avoid litigation for the use of material that it does not own and to earn a proﬁt. And yet this high-cost, low-revenue, litigation-prone10 startitfacts.biz/index.php?id=P968. stateofthenewsmedia.org/2006/narrative_networktv_audience.asp?cat=3&media=5. 7 washingtonpost.com/wp-srv/politics/polls/wat/archive/wat061900.htm. 8 usatoday.com/money/media/2006-05-08-newspaper-circulation_x.htm. 9 doublefusion.com/advertisers-media-buyers/market-overview.php. 10 Not surprisingly, Viacom and others have indeed recently sued Google for copyright infringement by YouTube; for example, see the article in the New York Times, nytimes.com/2007/03/14/business/14viacom.web.html?ex=1331524800&en=e9679584c2827550&ei=5088&. For an interesting analysis of the basis of litigation and of YouTube’s prospects, you can view the video posted on YouTube that explains the case, youtube.com/watch?v=rr9SQ4qkMMk&eurl=.
up with an uncertain business model was recently acquired by Google for $1.65 billion, based on the expectation that as a social network website it would produce signiﬁcant revenues going forward. • MySpace was acquired by News Corp for $580 million, with a similar and similarly vague value proposition. Several months later, there was little data to support the value of MySpace, either to News Corp or anyone else.11 • Microsoft acquired 1.6% of Facebook for $240 million, a very reasonable sum given its potential use to Microsoft, although one that places the improbably valuation of $ 15 billion upon Facebook. We explore this in Section 7.5. The Economist is especially harsh in their assessment, arguing that while online social networks will be everywhere, it is not at all clear that they represent viable businesses or viable business models. It is not clear to what extent revenues on these websites are expected to come from membership fees, corporate sponsorship of messages, traditional advertising, or other business models. News Corp's initial press release about their purchase of MySpace emphasizes “ … their commitment to expand its Internet presence by offering a deeper, richer online experience for its millions of users.” Google's oddly similar initial press release describing their rationale for buying YouTube discusses “… a successful brand and passionate community. The combined companies will focus on providing a better, more comprehensive experience to users…” Neither provided much information about what is actually planned, what services will be provided, or who would pay for them or why. The stated hope for both acquisitions is that advertising revenue will pay for and thus provide justiﬁcation support for the prices paid for each site. But how this will happen is left vague in both press releases since neither News Corp nor Google really knows how this anticipated advertising revenue would be generated. Since users do not go to those sites for advertising-based content, the assumed huge advertising revenues may not materialize. Likewise, the assumption that social networks are just new media, through which old advertising strategies can be executed, seems questionable at best. Google has experimented with bottom banners during the initial few seconds of a YouTube video , but these do not appear to have been widely implemented. Google continues to plan wider use of advertising on YouTube , but at present advertising on YouTube still seems quite limited. Upbeat reports continue to be written about Google and Hulu, but the question of how to convert these into viable businesses relying upon demonstrably effective advertising remains unanswered. Likewise, there are some indications that YouTube is doing a better job of pulling unauthorized use of copyright material, and that YouTube and the TV networks are exploring revenue sharing models, but until the revenue from YouTube advertising becomes signiﬁcant it will be difﬁcult to assess how these revenue sharing models will work out. At present the purchases still have an all-too-familiar resemblance to the wild escapades of the dotcom era when fortunes were paid for sites with no revenue stream and no strategy for generating revenue on a regular and sustained basis. This is not to say that these websites are without value, or even that they are without considerable value; it merely states that no compelling value proposition has yet been disclosed. It is especially true that directly replacing mass media advertising revenues with advertising revenues from electronic social networks seems implausible over the long term, despite whatever early revenue successes may initially be observed. If viewers discount commercial content displayed in television ads it is likely that they will soon come to discount commercial content displayed on websites like
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
Table 1 Classiﬁcation of ESN promotional revenue sources and the likelihood of their ultimate success.
YouTube. If we would be outraged to ﬁnd that a friend recommended a car or a dealership to us because he had been paid to do so, it is likely that we would be just as outraged to learn that recommendations from friends in an ESN were thinly disguised commercial messages. 6. How we use electronic social networks There are several different forms of electronic social networks. Some ESNs provide a professional service, like LinkedIn. Some form an electronic community dedicated to sharing information on a speciﬁc category of goods or services, such as user-generated reviews to rate travel service providers, restaurants, or other consumer purchases, or user-generated content, like Flickr or YouTube. Still others offer true, general-purpose social interaction, without virtual reality; these are the asynchronous interaction systems like MySpace and Facebook. Finally, there are the virtual reality experiences, like Second Life or World of WarCraft. Each can be used in one or more different ways; we focus here and in the subsequent sections on the individual's intent
when using them, rather than on the technology or the nature of the user interface. We see six principal forms of usage for electronic social networks, which closely mirror the way we use traditional social networks: 1. Pure fantasy and entertainment — Second Life and World of WarCraft are by far the best-known examples of massively multiuser domains dedicated to simultaneous fantasies and game playing. (This corresponds to the ﬁrst and second rows in Table 1, discussed in Section 10.) 2. Social interaction — MySpace and Facebook are by far the bestknown examples of online virtual communities, where users can post their homepages, describe their activities, link to and communicate with online friends, and coordinate activities within sub-communities. (This corresponds to row 5.) 3. Social interaction aimed principally at exchanging information — This includes sharing reviews on Amazon, a shopping site, or posting reviews on sites that exist only to provide a shared source of
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
information for the community that posts content. TripAdvisor supports travelers (tripadvisor.com), Chowhound supports and informs foodies (chowhound.com), and Ratebeer (ratebeer. com) does the same for serious beer drinkers. (This corresponds to row 4.) 4. Information aimed at exchanging information among professionals — This can be information for job search or the pursuit of professional objectives (LinkedIn), in support of professional associations, or in support of working relationships between companies and users, such as physicians and pharmaceutical companies. (This corresponds to row 3.) 5. Information and Support — Newly divorced women, pregnant women, men recently diagnosed with cancer, allergy sufferers, bereaved parents, and other groups increasingly use the internet to communicate with each other, share information, and offer mutual support. (This also corresponds to row 4.) 6. User-controlled content and entertainment — YouTube (youtube. com) and Flickr (ﬂickr.com) are among the best known examples. This is just another form of media, albeit one with less formal, more user-generated content. (This is not included in the table.) There is no reason to suspect that this list is mutually exclusive; for example, an individual may use Flickr to share content, for social interaction, and to exchange information and learn useful photographic techniques. Likewise, this list may not be exhaustive, and as the media richness available online increases , new uses of social networks may emerge. 7. The value proposition of social networking Just as we see multiple reasons why individuals use social networks, we see multiple mechanisms through which their owners might attempt to monetize them. 1. Using social networks for ethnographic observation — Fantasy social networks like Second Life allow users to do whatever they wish, limited solely by their imagination. Observing what they choose to do and not to do may provide insights into their wants and needs, cravings and longings, allowing observers to develop breakthrough innovations in products and services. 2. Using social networks to collect usage fees — These can range from monthly subscriptions for World of WarCraft to rent on unreal real estate in Second Life, or for add-on software that increases the capabilities or improves the appearance of the user's avatar. There have been attempts to use Second Life to host virtual and mixed reality meetings. However, given the current limitations of the technology, it is unclear why a virtual meeting would be preferable to current forms of teleconfererencing? 3. Revenues from selling through online social networks — Websites can earn money either through sales (as Amazon does with the books that it sells) or through linkages with websites that sell (as Amazon does with merchants who are part of the extended Amazon market). 4. Using social networks to replace traditional advertising — Websites can earn revenues from ads, near-ads, and concealed ads. 5. Using social networks to create synergies for other online resources — If Facebook can be linked to Microsoft's search mechanism, so that MSN's search returns items that are most likely to be relevant based on the searches conducted by my friends, then MSN search becomes far more useful to me than Google, and Microsoft becomes superbly repositioned in its battle with Google for internet supremacy. Although we attempt to make a sharp distinction between a website earning revenues by selling advertising space or charging fees for participation, the distinction is less clear than this suggests. For example, Second Life may charge rent, a form of usage fee, to a commercial user, who in turn attempts to monetize this investment by
using the space for advertising or billboards. That is, different participants in the same website simultaneously may be attempting to exploit different techniques for monetization. Of course, if it proves impossible to sell ads, then monetization based on renting space for running ads will ultimately fail as well. We address each of the ﬁve proposed monetization techniques in turn. 7.1. Using social networks for ethnographic observation Online social networks do allow individuals to tailor their avatars to have any appearance they wish, from real to idealized, and from idealized to fantastic and unreal. But what does observing these avatars tell us? After a considerable amount of time viewing Flickr postings from Second Life participants, the principal things we could learn about humans' idealized appearance, and about human behavior more generally, would appear to be the following: • When people can choose their appearance, most of them opt for well-known signs of youth, beauty, and health. Perfectly symmetric features, clear skin without warts or blemishes, and white teeth are a plus. • Indeed, people choose appearances that resemble sexual caricatures. • Flagrant displays and sexual signaling, beyond anything currently permissible in our ﬁrst lives, do get attention. When that fails, personal appearance modiﬁcations that cannot be employed in our ﬁrst lives, such as well-groomed wings and animal ears are effective as well. It is hard to imagine charging for these insights; some are already self-evident to marketers in ﬁrst life, while others, particularly wings and ears, are interesting solely because they cannot be applied here in our physical world. Likewise, what can we learn by observing avatars' activities and their behavior? Most of us enjoy sunbathing with a book on a quiet beach. Any of us lucky enough to spend a December evening in a hot tub in Tahoe, with warm water, frigid air, good friends, and a billion stars will always remember the experience. Similarly, memories of a tranquil sail on a quiet lake with a gentle breeze provide much joy. It is not yet clear to most observers why anyone would choose to pseudoexperience any of these in Second Life, since the experiences are at best a pale copy, but that is not the point. Quickly scanning Second Life posts on Flickr teaches us that many activities that are deemed enjoyable in our real ﬁrst life existence, and especially activities that appear to confer high status or to be exclusive, remain popular in Second Life as well. We observe avatars eating, dancing in exclusive clubs, walking through the Armani store, participating in beauty pageants, or skydiving. At present, at least, observing what people are doing in Second Life, as a surrogate means of determining what they wish they were doing in their ﬁrst life, seems to provide very little new insight into consumers' preferences. And yet, we are forced to note that communities have been very effective in designing modiﬁcations to existing products and in communicating not only their preferences but also their designs to the ﬁrms that provided the original products. MIT Professor Eric von Hippel has documented user-designed enhancements in a range of industries, from software to climbing equipment, and from consumer products to surgical implements [36,37]. Arakji and Lang have documented the extent to which innovation in computer games is determined by the communities of online gamers, not solely by the game companies . Feedback from users and “feedback-inﬂuenced design”12 have become signiﬁcant in a wide array of consumer products, including automobiles (GM's H3 Hummer), fast food (Taco Bell), health and beauty aids (Unilever), and insurance (State Farm).
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
While observing Second Life fantasies may not be the best way to capture these insights, it seems inevitable that some means of harnessing online social networks for product design will occur. We have already observed use of emerging trends in YouTube to inform selection of targets of opportunity for innovative television programming, and perhaps trends in Second Life (e.g., the emergence of Furries13) or in Facebook may likewise suggest product or service offerings here in our ﬁrst life. 7.2. Using social networks to generate usage fees It seems self-evident that many people around the world enjoy participating in Second Life or competing in World of Warcraft. It is not clear how large the revenues from Second Life subscription fees, unreal estate rentals and purchases, fees for avatar enhancements, or fees for additional avatar capabilities will become. Revenues from the online gaming segment of social networking14 are growing quickly, now exceed those of Hollywood15 and are half those of cable television,16 so the prospects for usage fees from successful social networking websites would appear to be considerable. Jaron Lanier  argues that ultimately the net cannot rely solely on advertising and piracy, and that ultimately creative talent will be paid for online access to content. Similarly, it seems plausible that when users ﬁnd that their online communities are truly valuable they will pay membership fees, as they currently do to participate in religious organizations, social clubs, alumni clubs, and professional associations. However, charging for content has to date enjoyed only limited success.17 Both the New York Times and The Wall Street Journal have recently dropped fees for viewing their content online. Indeed, as long as almost everything is available without charge, and subsidized through advertising, charging for content will remain extremely difﬁcult. 7.3. Using social networks to sell Social networks obviously already have been harnessed to sell, or to inﬂuence sales, online. Amazon is primarily a shopping site, but its online reviews, posted by its community of shoppers, are a valuable resource upon which other shoppers depend. There is no doubt that materials, both physical goods and information goods, can be sold online. The question that we seek to address is whether or not ESNs can be harnessed as retailing tools. Some of the claims do appear patently absurd. Faculty do not use the classroom to sell products for their own commercial gain or because they are paid to do so, nor do their students. Few of us have explicitly tried to sell our products at social gatherings, family gatherings, religious services, or country clubs. We do observe what our friends, colleagues, and peers eat, drink, wear, or drive, and we do emulate each other. We exchange recommendations. But social networks are not principally, or even secondarily, retailing opportunities. This viewpoint is not universally shared. The age groups that frequent Facebook, MySpace, and YouTube frequently “buy stuff online”, but they cannot shop online at these three websites. As USA Today notes, “And that means that there may be billions of dollars being left on the table by the major social networking websites, according to an eyeopening survey to be released today by the American Marketing Association.”18 Of course, we cannot buy stuff online when we check USAir ﬂights (usairways.com/awa/) or check out TV offerings using AOL (tvlistings.aol.com/) or read the New York Times (nytimes.com/) or check stock prices at the New York Stock Exchange (nyse.com/). Does
xahlee.org/sl/fur.html. businessweek.com/innovate/content/aug2007/id20070813_120384.htm. breitbart.com/article.php?id=060309211938.c3imi3s8&show_article=1. 16 industry.tekrati.com/research/9602/. 17 news.cnet.com/8301-10784_3-9816100-7.html. 18 See “Survey, Social-network sites could also lure shoppers,” USA Today, p 1-B, November 24, 2006.
14 15 13
this suggest that all of these websites are also leaving billions of dollars on the table? How “eye-opening” would that be? While online sales are important, virtual reality sales efforts seem misguided. At present the code of ethics of the Word of Mouth Marketing Association does seem to be limiting sock puppeting and ads masquerading as friends' recommendations, and most commercial activities in Second Life are through transparently corporate activities like virtual stores.19 The present limitations of selling through Second Life become clear as soon as virtual stores' websites are examined. Yes, it is probably possible to buy a dress or a hat for an avatar on Second Life. Buying clothes for yourself is more difﬁcult; you can't sense the wonderful feel of a hand-ﬁnished suit at an online Armani store, nor can you judge precisely how it looks on you with your skin coloration, and of course you cannot get the ﬁnal alterations done by an expert Armani tailor. Even people who routinely buy their electronic equipment or book their travel online are reluctant to replicate the old shopping experience by going to Second Life. Why would anyone walk through a simulated store to see a simulated 160 GB iPod 6G when he can buy one at Amazon or B&H so much faster? Why would anyone go to a simulated travel agency to see a simulated agent type on a simulated screen when she can book so much faster simply by going to Orbitz or to the airline's own website? While attempting to sell in Second Life is probably ineffective, and attempting to sell to MySpace or Facebook friends would probably be inappropriate, attempting to sell in themed games like World of WarCraft is positively loathed. Selling will not be the principal revenue source of most electronic social networks. Of course, this does not mean that online sales are unimportant, merely that hijacking massively multi-participant online worlds as retailing opportunities is unlikely to succeed. An interesting subcategory of sales has emerged, that of selling virtual items, like gaming accessories or clothing for avatars, online in a virtual world . It is not yet clear why people buy these virtual items or how large the market can become. 7.4. Using social networks to replace traditional advertising If social networks generate sufﬁcient trafﬁc, it is tempting to view them as just new media, where the old model of pushing content out on captive viewers remains viable. Social networks are not media but milieux, places where people freely congregate, freely exchange information, freely entertain and inform themselves, and freely leave. It does seem clear that online reviews, internet referrals from friends, and even posted content can sell. Amazon has found that posting short cuts from the CDs it sells and the ability to “search inside” books it offers both increase sales. Reducing uncertainty increases consumers' conﬁdence and increases their willingness to pay for unfamiliar items. As importantly, the music group Arctic Monkeys has captured the attention of marketing executives around the world by launching itself successfully through its MySpace posting when no record company was interested in offering them a contract. But what are the lessons from the success of Arctic Monkeys and what are the questions that remain unanswered? • Is the success of Arctic Monkeys still tied to MySpace in a way that MySpace can monetize once the success of the group has been assured? • Can the success of Arctic Monkeys be replicated by allowing other popular music groups to post on MySpace and then charging them for referrals or sales or click-throughs to their own website for sales? • Can the success of Arctic Monkeys be replicated and accelerated by sending paid promotional messages, sock puppeting spam, to other MySpace users, to convince them to purchase the music of other groups that also do not have contracts?
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
The answer to the ﬁrst question would appear to be no; Arctic Monkeys would appear to be free of its initial dependence on MySpace. Over time, the origin of web trafﬁc on ArticMonkeys.com changed, from trafﬁc that originated on MySpace, to trafﬁc that originated through search, and ﬁnally to trafﬁc that went directly to the website by keying in its ArcticMonkeys.com URL. This is tracked by hitwise.com. Can other groups replicate Arctic Monkeys success? In principle, of course they can, but in practice it should prove extremely difﬁcult. Most great new groups will be offered recording contracts, and most great new groups will want them and will accept them. There will be some groups that will not get contracts and will have to promote themselves through MySpace, but we suspect that their ability to achieve such mass-market successes will be rare. Finally, can success be accelerated, for a fee, by using MySpace to send a large number of paid referrals, with pseudo-friends sending paid pseudo-endorsements? Of course, this violates the code of ethics of the Word of Mouth Marketing Association. As importantly, it is likely to be detected and rejected, and thus it is likely to be ineffective. Once again, social networks are meeting places or milieux, where participation is voluntary, not broadcast media where messages can be forced out upon millions of passive viewers. While the success of Arctic Monkeys indicates the power of social networks and word of mouth, it does not suggest that social networks can be hijacked, used to distribute advertising, and thus easily monetized. There are indeed some greatly entertaining and powerfully engaging corporate-funded messages on YouTube: • Dove soap's self-esteem campaign: http://youtube.com/watch? v=rJ9IPZLPGr4 • BMW Madonna “Star” video http://www.youtube.com/watch? v=e3R9IpAQ6xo The sales impact, and indeed the takeaways from these messages, both remain unclear. Dove, through research at OgilvyOne, learned that many women do not, and cannot, aspire to the anorexic images of supermodels with airbrushed skin and perfectly symmetrical features, and learned that these women feel that products advertised using those supermodels do not speak to them. By using women with more average ﬁgures and facial features in their advertisements, Dove hopes to capture those segments of the market that do not respond to idealized models. The campaign on TV and in print media has been successful but it is uncertain whether the YouTube extension will add to that success. YouTube is not seen by typical users as a place to go to look for presentational advertising messages, and most of the intended Dove audience will in all likelihood not even see the Dove message on YouTube, much less respond to it. As we continue to try to port our ﬁrst life advertising over to second life experiences, billboards have made their appearance in Second Life. Billboards along the roadside have long been considered eyesores, and are banned in some communities; where they exist motorists drive past them on their way somewhere else. Most users teleport so quickly from one place in Second Life to another that billboards must be strategically (and annoyingly) located to have any impact; apparently, when we can't ignore them, we resent them, and most of the blogosphere's reactions to Second life billboards at present appear to be negative.20 Unlike media advertisements, which captive audiences had to view, ESN messages are viewed only if the user wishes to do so. If ESN messages are seen as intrusive, as the Beacon venture between Microsoft and Facebook was, the reaction is overwhelmingly negative.21 Monetizing media audiences was much easier than mone-
tizing milieux participation. Again, this view is not universally held . 7.5. Using social networks to create synergies for other online resources Linkages between review sites and sale sites may be the most powerful and least intrusive way to harness social networking sites. While Facebook and MySpace have been slow to create linkages to sites that actually sell, others have done this effectively, with both implicit and explicit linkages. TripAdvisor was originally owned by Interactive Corp, which also owned hotels.com, and it made hotels. com enormously more valuable; travelers who were abandoning their traditional agencies for online search needed a reliable source of information, and TripAdvisor provided it without charge. OpenTable (opentable.com) now has a relationship with Zagat. The relationship between the Beer Yard (beeryard.com) and Ratebeer has been a brilliant success for the Beer Yard, though it does not yet appear to have done much for Ratebeer. A beer lover can go to the Beer Yard's website and with one click can ﬁnd out what beers have recently arrived. With a second click the potential customer can get directed to Ratebeer and can read reviews from numerous other reviewers, decide if he wants the beer, and if so he can call and reserve a case. This has been great for sales of craft brews at the Beer Yard, and it is probably a harbinger of relationships between sellers and independent user-generated content reviewing sites. We expect to see far more of this in the future, if indeed a way can be found to make it work for both parties, without compromising the trust users have in their rating website. Microsoft's recent investment in Facebook could still turn into the most subtle attempt to create synergies that we have seen so far. Google is, of course, Microsoft's most feared opponent, and at present 98% of Google's revenues come from paid search, search results returned to the user on the top of the results page, not because they are most relevant to the user but because the companies listed have paid top dollars for top billing. At some point users will be aware of this, will skip or discount paid search, and Google will need a better business model. Microsoft's investment in Facebook allows it to develop that better business model almost immediately. With Facebook linkages Microsoft can tell someone what his or her Facebook friends found to be relevant when they searched. Microsoft can also, with the friends' permission, tell the searcher what those friends thought of the products or services that he or she is presently considering. While this would not, in and of itself generate revenues for Microsoft, it would position MSN superbly to outﬂank Google. 8. The new 4 Ps of social network marketing It would be useful, when trying to predict what it takes to attract a following to a social networking website, to observe what these websites have in common. It appears that all social networking websites that attract a following offer at least two of the following 4 Ps: • Personal — They are personally relevant. They resonate with a real interest of their participants. Whether the interest group is all newly admitted freshmen of the University of Chicago, or all serious students of beer, or those interested in learning how to deal with a new divorce or how to cope with a newly diagnosed disease, or those interested in learning how to plan a wedding, community websites must provide resonance for their viewers and users. • Participatory — Ultimately, the best social networking websites allow participation. The audience at a play is not a social network, but an acting troupe that selects its plays, understudies and improvises for each other, and then performs is most deﬁnitely a strongly interactive social network. When ESNs encourage interaction and active participation they go beyond serving as personalized
secondlifeherald.com/slh/2007/03/big_business_ad.html. See gigaom.com/2007/11/06/facebook-beacon-privacy-issues/ or publishing2.com/ 2007/12/01/facebook-beacon-a-cautionary-tale-about-new-media-monopolies/.
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
traditional media they become more functionally equivalent to social networks and more valuable to their members. • Physical — Many social networking websites appear to offer the possibility of a transition to physical meetings in our real world. Obviously that is central to the mission of all dating websites. But websites as diverse as Ratebeer, Flickr, and World of Warcraft seem to offer their participants opportunities to meet outside the cyberspace conﬁnes of the websites. In other contexts, particularly with individuals who have adopted a persona and an avatar very different from their real world selves following up with meetings in person may prove far less attractive. • Plausible — The artiﬁcial worlds of Second Life and World of Warcraft offer physical constraints and limitations different from those we experience on earth. Participation and performance require at least some rules, to make them sufﬁciently plausible to allow users to function. When some gamers started using death as means of free teleportation the rules of death were changed, but World of Warcraft felt compelled to defend the new rules and explain why the change was necessary.22 9. Major source of uncertainty
10. Conclusions 10.1. Forces at work Any attempt to predict the future of advertising must understand both consumer behaviors with regard to shopping and their behaviors in their use of the internet. Consumers use the net to see what they want and do what they want, when they want it. The net will not support entrapment, and consumers who are escaping ads through TiVo are unlikely to become attracted to full time advertisement programming. The modern consumer views the net as transparent, and community postings and the recommendations of friends will outweigh paid placements, and paid reviews. Placements of products in movies or in video games are likely to be ignored; the presence of advertisements directly into movies or video games will be resented. Virtual reality websites simply are not ideal for encouraging online consumption or for encouraging consumption in physical space. It's meaningless to drink a beer online, and pointless to attempt to inﬂuence my selection of a beer by drinking one online with me. I can't taste anything online, and I know that you can't either.23 10.2. Predictions
While mastering the new 4 Ps may ensure trafﬁc, this by no means guarantees that trafﬁc can be monetized or that the website can produce revenues or proﬁts. We have several sets of questions: • Can any social networking website attempt to commercialize trust without destroying that trust? Can websites charge for advertising, or receive payments for positive recommendations, without destroying the trust of its users and becoming seen as just another source of paid messages rather than of real conversations of and two-way communications between customers and ﬁrms? In brief, can trust be monetized, or does the mere attempt to get paid for trust destroy it? • Can any social networking website attempt to steer trafﬁc without destroying trust? Can it attempt to control interactions to commercialize them without destroying trust? • Do social networking websites actually move business? Do they deserve to be paid, quite separate from whether or not any mechanism exists to monetize trust? Questions in the ﬁrst set remain unanswered and they pose the biggest challenge to commercializing and monetizing News Corp's investment in MySpace. Questions in the second set are closely related; that is, if websites can create business just by steering users, and if users will tolerate content that attempts to steer them, then MySpace, Facebook, and Flickr may have huge untapped value. It is plausible, but uncertain, that in many instances users would abandon websites that were populated by pseudo-users who were little more than paid corporate shills. And yet, we have consistently failed to see patterns in the use of new technology and its transformation of society. The impact of the telephone and the elevator on the nature of work, allowing towering skyscrapers and densely populated urban centers is obvious only in hindsight. Likewise, the impact of the internal combustion engine and the automatic transmission, combined with inexpensive gasoline, in hollowing out many of those American cities, was obvious only after it had occurred. The interaction of VisiCalc with the Apple II contributed greatly to the successful launch of personal computing. It is at least possible that the interaction of electronic social networks with some other technology, perhaps mobile computing and mobile telephony, will be equally transformational in affecting the way we learn about products and conduct commercial transactions.
The internet already does and will continue to inﬂuence sales. • Yes, consumers inﬂuence each other through posted content. • Yes, it is possible to sell online. But hijacking social networking websites does not appear to be the most effective way to do so. Some phenomena are too complex to predict right now. If a truly stunning avatar were to be seen carrying a great new product, or wearing a bold new style, that might indeed inﬂuence styles in the real physical world. Consumer media viewing behavior, consumption behavior, receptivity to advertising, and use of information in purchases are all changing. The changes are profound, and they interact in complex ways that confound easy analysis. The most tempting response among advertisers and among their agencies is to assume that the most signiﬁcant changes are in media choices and viewing behavior, which require the smallest response — change the venue in which ads appear, and change their format as dictated by the new media. While we do not yet know enough to determine if a change in advertising venue will be sufﬁcient to restore the power of advertising, we suspect that it will not be, and that the high valuations placed on early social networking websites may be irrationally high, even as hedging bets. One thing, at least, is certain: research will continue . 10.3. Classiﬁcation of social network promotions Our treatment of internet advertising is less optimistic than that of many other authors, and may require further treatment. For the purposes of this paper, we focus on a range of forms of promotional informational activities, those based on providing information, as opposed to those providing price discounts, or visibility by sponsoring NASCAR racing teams, or visibility by distributing free samples at sporting events. Within
23 It is certainly true that I cannot at present enjoy a beer online. It is remotely possible that I may be able to enjoy a beer and truly experience it online, or have a meaningful test drive of an automobile online at some time in the future. Philosophers from Descartes and Kant to the present have debated the meaning and even the reality of experience. Science ﬁction writers made signiﬁcant contributions to the debate, with novels like Neuromancer  and Snow Crash . The science ﬁction writers were temporarily trumped by philosophers again, with the Brain in a Bottle hypothesis, which argued that possibly nothing was real, or that we would have no way of knowing if it were real or not. The direct applicability of philosophy to the analysis of perception of beer can be reviewed in . Sufﬁce it to say that at least for now we can sample music and video online, but we cannot sample beverages or test drive automobiles, and that the value of experiencing consumer goods online was not an initial motivation for the acquisition of YouTube or MySpace.
E.K. Clemons / Decision Support Systems 48 (2009) 46–56
informational promotional activities, we divide the collection of actions into advertising and sponsored content. 1. Advertising is the push-based provision of short messages, without user request; it is generally embedded in the media content that the user has chosen to view, such as newspaper pages or television shows. Increasingly, it can be tailored to user proﬁles, as with popups and banners that are prompted by what the user is viewing, but it is generally not provided in response to users' requests for information on products or services. 2. Sponsored content is pull-based, in that it is in response to users' requests for information, and is almost always more ﬁnely tuned to individual requests. This paper assumes that advertising is losing its effectiveness, for two reasons: • Ads are losing their viewers; we no longer have patience for commercials embedded in our TV programming and TiVo over it, and we no longer have patience for pop-ups and banners and employ software to block them whenever possible. To the extent that this is true, moving advertisements to media that provide even greater ﬂexibility and control to viewers will not improve the visibility of advertisements. As importantly, most forms of mass media are themselves losing viewers, compounding the problems faced by advertisers. There is extensive support for this, across all media. • Ads have lost their credibility and much of their ability to inﬂuence consumer opinions and consumer decisions; they don't view them because, quite simply, they no longer trust them. While there is less support for this, it appears plausible, and for the purposes of this paper we accept it as well. To the extent that this is true, transferring advertisements to alternative media, or ﬁnding alternative means for their distribution, will have but limited impact. The paper makes less extreme assumptions concerning consumers' response to sponsored content that they have actually requested. It assumes that ﬁrms have the opportunity to regain viewers if content is requested rather than pushed out, and that ﬁrms have the opportunity to regain some credibility with sponsored content. Still, there is some support for the fact that content provided by ﬁrms to describe their own goods and services may have less credibility than content provided by independent rating services. We divide ESNs into four coarse categories: 1. Some ESNs are multi-media virtual reality settings, in which users simulate interactions as if they coexisted in a shared virtual world. The degree of immersion in and intensity of these virtual worlds approaches that of reality as technology improves. At present, most interactions in virtual worlds are social, rather than commercial or professional in nature. Second Life would be the best-known example of the milieux. 2. Some ESNs are multi-media virtual domains, but do not attempt to mimic reality. These are often not general purpose alternative worlds, but rather dedicated to a single form of interaction. Games like World of Warcraft are the best-known examples. 3. Some ESNs are dedicated support systems for professional use. Some professional associations have ESNs, and some networks like LinkedIn serve this function as well. 4. Some ESNs are used primarily for ongoing social communications, and while they may permit sharing of audio or video, they are primarily asynchronous textual interactions, like MySpace or Facebook. Increasingly, they are using them to establish their own online communities within a larger network; universities have taken the lead on this but many corporations are experimenting as well. Again, this paper assumes that the market for push-based advertising will be extremely limited, and that the effectiveness of such advertising will be very limited, regardless of how much money is
currently spent on it. Likewise, it does not rule out roles for sponsored content, embedded as appropriate, although the use of this is minimal today, and although we consider this to be promotion but not truly advertising. Indeed, after wrestling with how to monetize community trust without compromising users' belief in their honesty, integrity, and total commitment to users' needs for over two years, the designers of First Wives World decided that they would have no advertising. Instead they feature an advice box, in which advertisers and sponsors can provide information on any subject the user wants, but only in response to users' speciﬁc requests; in addition, other users' assessments and reviews will also be provided along with the sponsored content. A preliminary classiﬁcation is shown in Table 1. Material is indicated as either in the zone of probable failure, a zone of strategic uncertainty, or a zone of probable success. Social search, in the lower left corner, is crossed out; it is search, not advertising or promotion. 10.4. Directions for future research The revenue from text-based ads, banners, sidebars, and other content placed around web pages, is presently very high. Although ﬁrms can measure some elements of consumer interaction, such as views or impressions, it is not yet clear that the strategic or tactical impact is high. The actual value received by advertisers needs more careful measurement. We believe that sponsored content will have higher impact than ads. Firms will beneﬁt more from explaining the true advantages of their products than from trying to convince potential customers through traditional shots of consumers using the products and praising them. Still, Ariely's work suggests that the same review will have greater impact if posted on a professional rating site or a community content site than on the manufacturer's site. This needs more conclusive study. References
 R.D. Alba, C. Kadushin, The intersection of social circles: a new measure of social proximity, Sociological Methods and Research 5 (1996) 77–102.  C. Allen, D. Kania, B. Yaeckel, One-to-One Web Marketing: Build a Relationship Marketing Strategy One Customer at a Time, Second editionJohn Wiley & Sons, 2001.  R.Y. Arakji, K.R. Lang, Digital consumer networks and producer–consumer collaboration: innovation and product development in the digital entertainment industry, Journal of Management Information Systems 24 (2) (2006) 195–219.  D. Ariely, Predictably Irrational: the Hidden Forces That Shape Our Decisions, HarperCollins, 2008.  J.A. Barnes, Social networks, Addison-Wesley Module in Anthropology 26 (1972) 1–29.  S. Barnes, Virtual worlds as a medium for advertising, ACM SIGMIS Database 38 (4) (2007) 45–55.  S.P. Borgatti, M. Everett, Models of core/periphery structures, Social Networks 21 (1999) 375–395.  D.A. Bray, B.R. Konsynski, Virtual worlds: multi-disciplinary research opportunities, ACM SIGMIS Database 38 (4) (2007) 17–25.  E.K. Clemons, Monetizing the internet: surely there must be something other than advertising, Proceedings, 42nd International Hawaii Conference on System Sciences, 2009.  Clemons EK, Barnett S, Magdoff JM and Clemons JC. Managing the “Instant Messaging” Generation in the Work Force, in Future of Networks, J. Wind and P. Kleindorfer, (eds.), forthcoming.  E.K. Clemons, G. Gao, Consumer informedness and diverse consumer purchasing behaviors: traditional mass-market, trading down, and trading out into the long tail, Electronic Commerce Research and Applications 7 (1) (2008) 3–17.  E.K. Clemons, G. Gao, L. Hitt, When online reviews meet hyperdifferentiation: a study of the craft beer industry, Journal of Management Information Systems 23 (2) (2006) 149–171.  E.K. Clemons, R. Spitler, FT survey on mastering innovation, the new language of consumer behavior, Financial Times (October 7 2004) 4–5.  E.K. Clemons, B.W.R. Weber, Turmoil, transparency, and tea: valuing the impact of IT on London's stock exchange, in: R. Kauffman, R. Banker, M. Mahmood (Eds.), The Strategic and Economic Value of Information Technology Investments, Idea Group Publishing, September 1992, pp. 202–235.  C. Dellarocas, The digitization of word-of-mouth: promise and challenges of online reputation systems, Management Science 49 (10) (2003) 1407–1424.
E.K. Clemons / Decision Support Systems 48 (2009) 46–56  E. Von Hippel, Lead-users: a source of novel product concepts, Management Science 32 (7) (1986) 791–805.  E. Von Hippel, Democratizing Innovation, MIT Press, 2005.  J. Wais, E.K. Clemons, mobile, 2008.  D. Watts, J. Peretti, Viral marketing for the real world, Harvard Business Review (May 2007) 104–112.  W.F. Whyte, Street Corner Society: the Social Structure of an Italian Slum, 4th editionUniversity of Chicago Press, 1993.  J. Wind, V. Mahajan, Digital Marketing: Global Strategies from the World's Leading Experts, John Wiley & Sons, 2000.
 A.R. Dennis, S.T. Kinney, Testing media richness theory in the new media: the effects of cues, feedback, and task equivocality, Information Systems Research 9 (3) (1998) 256–274.  D. Fossey, Gorillas in the Mist, Mariner Books, 2000.  W. Gibson, Neuromancer, Ace, 1984.  J. Goodal, My Life with the Chimpanzees, Aladdin, 1996.  J. Goodal, In the Shadow of Man, Mariner Books, 2000.  Y. Guo, S. Barnes, Why people buy virtual items in virtual worlds with real money, ACM SIGMIS Database 38 (4) (November 2007) 69–76.  S.D. Hales (Ed.), Beer and philosophy: the unexamined beer isn't worth drinking, Blackwell, 2007.  G. Hardin, The tragedy of the commons, Science 162 (1968) 1243–1248.  M. Helft, Google aims to make YouTube proﬁtable with ads, The New York Times, Nov. 20 2007.  M. Hughes, Buzzmarketing: “get people to talk about your stuff”, Portfolio, 2005.  E. Keller, J. Berry, The inﬂuentials, The Free Press, 2003.  J. Lanier, Pay me for my content, The New York Times, Nov. 20 2007.  M.W. Macy, J. Skvoretz, The evolution of trust and cooperation between strangers, American Sociological Review 63 (1998) 638–660.  B.P.S. Murti, S. Sarkar, The role of the management sciences in research on personalization, Management Science 49 (10) (October 2003) 1344–1362.  H. Rheingold, The Virtual Community: Homesteading on the Electronic Frontier, revised editionMIT Press, 2000.  J. Riedl, J. Konstan, E. Vrooman, Word of Mouse: The Marketing Power of Collaborative Filtering, Warner Business Books, 2002.  S. Senecal, J. Nantel, The inﬂuence of online product recommendations on consumers' online choices, Journal of Retailing 80 (2) (2004) 159–169.  N. Stephenson, Snow Crash, Bantam Books, 1992.  L. Story, YouTube's advertising coming-out party, The New York Times, Feb. 13 2008.  S. Strogatz, Sync: the Emerging Science of Spontaneous Order, Hyperion Books, 2003.
Eric K. Clemons is a professor at the Wharton School, University of Pennsylvania, where he has served since 1976. His visiting appointments include Harvard University, Cornell University, Hong Kong University of Science and Technology, and the Indian School of Business. He serves on the editorial boards of Journal of Management Information Systems, International Journal of Electronic Commerce, and eCommerce Research and Applications. His research specialties are in the areas of IT and business strategy, IT and Consumer Behavior, and IT and markets. More speciﬁcally, he studies strategies for proﬁting from internet applications, the changing role of internet advertising, the impact of technology on consumer behavior, managing the risks of strategic outsourcing, managing the risk of strategic IT implementations, and strategic implications of e-commerce for channel power and proﬁtability. He has an S.B. in Physics from MIT and an M.S. and Ph.D. in Operations Research from Cornell University.